Ownership Structure, Governance, and Innovation: Evidence from Italy
Total Page:16
File Type:pdf, Size:1020Kb
Ownership Structure, Governance, and Innovation: Evidence from Italy Raoul Minetti∗ Pierluigi Murro Monica Paiella Michigan State University Luiss University University of Naples Abstract This paper tests the impact of firms’ ownership structure on firms’ innovation decisions using a rich dataset of about 20,000 Italian manufacturers. After accounting for its possible endogeneity, we find that ownership concentration negatively affects the probability of innovation, especially by reducing firms’ R&D effort. The results suggest that conflicts of interest between large and minority shareholders are a determinant of the negative effect of ownership concentration on inno- vation. Moreover, risk aversion induced by lack of financial or industrial diversification appears to be an additional source of large shareholders’ reluctance to innovate. Once we distinguish across types of shareholders, we uncover some evidence that families support innovation more than financial institutions, but that the benefits of financial institutions for technological change increase with their equity stakes. Collectively, the findings support the view that the agency problems that affect firms in continental Europe differ from those in the United States, not only in static but also in technologically dynamic environments. JEL Codes: G32, O3 Keywords: Ownership, Agency Problems, Technological Change. 1Introduction Technological innovation is a key determinant of firms’ performance (OECD, 2010). Innovation affects the degree to which firms can enhance their productivity (Tellis, Prabhu and Chandy, 2007). It also influences their ability to penetrate into new markets, including foreign ones, or preserve their leadership as market incumbents (Geroski, 1995). Corporate governance can play a crucial role in firms’ ability to advance their technological frontier. For example, it is sometimes claimed that the public companies that are prevalent in the United States are suitable for promoting innovation because they can diversify the risk of innovation across a large number of investors (Aghion, Van Reenen and Zingales, 2009). By contrary, in recent years some policymakers have expressed concerns that family—oriented businesses, such as those typical of countries of continental Europe and East Asia, might be less prone to technological change (Onida, 2004). For example, families could be reluctant to abandon their traditional core business and venture into risky new activities. These arguments are however far from being established and the relation between firms’ ownership structure and innovation remains an open question. The objective of this paper is to shed new light on the effects of ownership on innovation exploiting a rich survey of over 20,000 Italian manufacturing firms conducted by the banking group Capitalia. The dataset ∗Corresponding author. Email: [email protected]. Address: Department of Economics, Michigan State University, 110 Marshall-Adams Hall, East Lansing, MI 48824-1038, USA. We wish to thank several seminar participants for helpful comments and conversations. All remaining errors are ours. 1 provides thorough information on firms’ innovation effort which is based directly on firms’ responses to survey questions. It also contains precise measures of firms’ ownership structure, such as the capital shares and the type of the largest equity—holders. Another advantage of our dataset is the availability of instruments for firms’ ownership structure. A challenge of any study on the relation between ownership and innovation is that unobservable factors can affect both the governance structure and innovation decisions. Moreover, reverse causality may also be an issue at play as technological progress itself can shape the ownership structure. Our strategy for tackling these issues is to employ information on past regulation of Italian local financial markets. It is believed that the ability of firms to obtain external finance (credit) is a key determinant of firms’ needs and incentives to open participation to new shareholders, issue new equity and go public.1 We thus employ information on the regulation of Italian local financial markets, introduced in the late 1930s, to capture exogenous restrictions on the availability of external finance and construct instruments for firms’ ownership structure. After accounting for the possible endogeneity of the ownership structure and controlling for a variety of factors that may also affect innovation, we find that ownership concentration negatively affects the probability that firms innovate: increasing the equity share held by the main shareholder reduces the likelihood of innovating by 15 percent, which corresponds to 40 percent of the mean likelihood to innovate. The negative effect weakens as ownership concentration increases, with the effect essentially tapering off for levels of ownership concentration above 70 percent of capital.2 This result holds regardless of whether we focus on the capital share of the main equity—holder or the two largest equity—holders. Furthermore, ownership concentration appears to be essentially neutral for total investment, signalling that its negative effect on innovation does not simply reflect a broader effect on the total volume of investment. When we subdivide the innovation process into its phases, we also find that ownership concentration depresses investment in R&D more than (proxies for) technology adoption. When we split the sample by size of the firm, we find that ownership concentration has a negative effect on innovation on large firms, and not on small firms, for whom the effect is not statistically significant. The effect is negative for firms in sectors where economies of scale are not important and for firms operating in ’traditional’ sectors. Ownership concentration does not seem to matter for high tech firms, for whom, presumably, innovation is a precondition for survival. Finally, it is worth mentioning that, throughout the analysis, we distinguish between product and process innovation. Most results mentioned so far refer to the former. In contrast, ownership concentration has no significant effect on process innovation, which is consistent with the hypothesis that the two respond to different objectives and may have different determinants (Cohen, 1995, and Cohen and Klepper, 1996). The analysis then turns to disentangle the mechanisms through which ownership concentration depresses innovation. Italy represents an ideal environment to study conflicts of interest between large and minority shareholders and their possible impact on innovation. In fact, the Italian corporate sector is characterized by a strong presence of individual owners that hold sizable equity stakes in companies, while institutional ownershipislessdiffused than in the United States. The results suggest that conflicts of interest between large and minority shareholders indeed contribute to the negative effect of ownership concentration on innovation. In fact, the presence of external managers (who are believed to safeguard minority shareholders’ interests against the opportunism of large shareholders) in the board of directors has a positive effect on the propensity 1 For example, the access to bank loans can impact investors’ ability to obtain funds to purchase equity (Caselli and Gennaioli, 2006), the need for firms to issue new equity or also the price at which new equity can be sold. 2 These results hold for product innovation, while ownership concentration appears to be neutral for process innovation. 2 to innovate. We also uncover evidence that attributing control to the main shareholder fosters innovation. This suggests that aligning cash flow rights with control rights may mitigate agency problems inside firms, for example because it induces controlling shareholders to fully internalize the effect of their decisions on the firm’s performance. In addition to this, we find that risk aversion induced by lack of financial or industrial diversification may be a further source of large shareholders’ reluctance to innovate. The last part of the paper examines whether the nature of the main shareholder plays a role in innovation decisions. Firms led by a family appear to be more prone to innovation than firms led by a financial institution. However, the benefits of ownership by financial institutions for innovation increase with their equity stake. There is growing empirical literature on the role of large shareholders in explaining corporate performance, but the evidence is mixed. Shleifer and Vishny (1986), McConnell and Servaes (1990) and Zingales (1995) find a strong positive relation between ownership concentration and corporate performance in the United States and some other developed economies and attribute this to large shareholders’ monitoring. For Japan, Kaplan and Minton (1994) show that firms with large shareholders are more likely to replace managers in response to poor performance. Yafeh and Yosha (1996) uncover evidence that large shareholders reduce discretionary spending by Japanese managers, such as advertising and entertainment expenses. Focusing on transition economies, Claessens (1997) identifies a positive relationship between ownership concentration and both voucher prices and stock market prices. Xu and Wang (1999) obtain similar evidence for a sample of listed Chinese companies. Finally, Barberis, Boycko, Shleifer and Tsukanova (1996) detect a positive relation between firm performance and ownership concentration in Russia. In contrast with these studies, other scholars find a negative effect of concentration on performance for non—U.S. firms.